WO2016056305A1 - 制御装置及び制御方法 - Google Patents

制御装置及び制御方法 Download PDF

Info

Publication number
WO2016056305A1
WO2016056305A1 PCT/JP2015/073055 JP2015073055W WO2016056305A1 WO 2016056305 A1 WO2016056305 A1 WO 2016056305A1 JP 2015073055 W JP2015073055 W JP 2015073055W WO 2016056305 A1 WO2016056305 A1 WO 2016056305A1
Authority
WO
WIPO (PCT)
Prior art keywords
value
control
standard deviation
unit
execution command
Prior art date
Application number
PCT/JP2015/073055
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
将哉 木村
英俊 池田
史雄 米谷
Original Assignee
三菱電機株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 三菱電機株式会社 filed Critical 三菱電機株式会社
Priority to US15/508,684 priority Critical patent/US10088813B2/en
Priority to CN201580054433.8A priority patent/CN106796416B/zh
Priority to JP2016552857A priority patent/JP6494648B2/ja
Priority to TW104132803A priority patent/TWI563356B/zh
Publication of WO2016056305A1 publication Critical patent/WO2016056305A1/ja

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0205—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system
    • G05B13/024—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system in which a parameter or coefficient is automatically adjusted to optimise the performance
    • G05B13/0245—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system in which a parameter or coefficient is automatically adjusted to optimise the performance not using a perturbation signal
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/39—Robotics, robotics to robotics hand
    • G05B2219/39241—Force and vibration control

Definitions

  • the present invention relates to a control device and a control method for controlling a target device by feedback control.
  • feedback control is used to achieve the required operation.
  • the feedback control it is necessary to adjust the control gain used to calculate the operation amount from the detected control amount.
  • One of the methods for adjusting the control gain appropriately is the limit cycle method.
  • control is performed by selecting and outputting one of two manipulated variables called binary control, thereby causing the controlled variable to vibrate at a constant or constant period, and based on the vibration waveform.
  • the dynamic characteristic of a control target is identified, and the control gain is calculated (Non-patent Document 1).
  • the vibration waveform is called a limit cycle waveform.
  • the limit cycle method in order to identify the dynamic characteristics of the controlled object, one of the operation amounts of the binary control is selected based on the sign of the control deviation, and the control amount is oscillated at a period that is considered constant or constant. It is necessary to generate a limit cycle waveform.
  • noise is included in the detected control amount, and when the positive or negative sign of the control deviation is reversed due to the noise, the on / off control operates regardless of the dynamic characteristics of the controlled object. In other words, chattering may occur and a limit cycle waveform having a constant or constant period may not be obtained, and the dynamic characteristics of the control target may not be identified.
  • Patent Document 1 describes that a limit cycle waveform having a constant or constant period is generated even when noise is included in a detected control amount. Patent Document 1 describes that chattering of binary control due to the influence of noise included in the detected control amount is prevented by providing a hysteresis characteristic in the determination of the sign of the control deviation. Patent Document 1 describes that the control gain is calculated by identifying the dynamic characteristics of the controlled object even when the detected control amount includes noise.
  • the vibration amplitude of the controlled variable becomes large, the controlled variable is saturated, and the limit cycle waveform is controlled.
  • the shape may be unrelated to the dynamic characteristics or the control target may be damaged, and the control gain cannot be calculated accurately.
  • Patent Document 1 when the limit cycle method is performed based on Patent Document 1, the operator can observe the magnitude of the noise included in the detected control amount in advance to remove the influence of the noise, and control It is necessary to set the magnitude of hysteresis so as not to excessively increase the amount of vibration amplitude. Therefore, Patent Document 1 has a problem that it takes a lot of time and labor to set an appropriate magnitude of hysteresis.
  • Patent Document 1 also has a problem of increasing the production cost of the control device.
  • the amount of noise included in the control amount increases as the control amount changes, or the control device
  • the amount of noise included in the controlled variable increases depending on the operating conditions of the electrical machines and electronic devices installed around the control target, and the influence of noise cannot be removed, and the control gain can be calculated accurately.
  • the operator determines the magnitude of the hysteresis based on the magnitude of the controlled variable, or based on the influence of the operating state of the electric machine and electronic equipment installed around the control device and the controlled object. Need to be changed.
  • the operator needs to check the magnitude of noise included in the control amount. Therefore, in Patent Document 1, it is necessary for the operator to continuously change the setting of the magnitude of the hysteresis, and there is a problem that requires a lot of time and labor.
  • the present invention has been made in view of the above, and provides a control device and a control method capable of accurately calculating a control gain by removing the influence of noise by appropriately setting the magnitude of hysteresis.
  • the purpose is to do.
  • the control device calculates a control deviation by subtracting a command value input from the outside and a control amount input from the control target device.
  • An operation amount is generated based on the subtractor, the control deviation and the control gain, a control calculation unit that outputs the operation amount, an adjustment execution command value indicating ON or OFF, and an adjustment execution command value
  • An adjustment execution command generation unit to output, and an adjustment time addition value is generated based on the control deviation and the hysteresis width setting value during a period in which the adjustment execution command value output from the adjustment execution command generation unit is on.
  • a standard deviation estimator that calculates a low frequency component removed signal from which a component has been removed and calculates a standard deviation estimated value that is an estimated value of the standard deviation of the low frequency component removed signal; and a hysteresis width based on the standard deviation estimated value
  • a hysteresis width calculation unit that calculates a calculation value and changes a hysteresis width setting value of the binary output unit to the hysteresis width calculation value.
  • control gain is accurately calculated by removing the influence of noise by appropriately setting the magnitude of the hysteresis.
  • Configuration diagram of control device and control target device The figure which shows typically operation
  • FIG. Configuration diagram of standard deviation estimation unit according to Embodiment 1 Configuration diagram of control device and control target device according to second embodiment
  • Configuration diagram of control device and control target device according to Embodiment 3
  • Configuration diagram of standard deviation estimation unit according to Embodiment 4 Configuration diagram of standard deviation estimation unit according to Embodiment 4
  • Configuration diagram of control device and control target device Configuration diagram of control device and control target device according to embodiment 5
  • FIG. 1 is a configuration diagram of the control device 100 and the control target device 10 according to the first embodiment
  • FIG. 2 is a diagram schematically illustrating the operation of the binary output unit 103 according to the first embodiment
  • FIG. 3 is a configuration diagram of a standard deviation estimation unit 106 according to Embodiment 1.
  • FIG. 1 is a configuration diagram of the control device 100 and the control target device 10 according to the first embodiment
  • FIG. 2 is a diagram schematically illustrating the operation of the binary output unit 103 according to the first embodiment
  • FIG. 3 is a configuration diagram of a standard deviation estimation unit 106 according to Embodiment 1.
  • FIG. 1 is a configuration diagram of the control device 100 and the control target device 10 according to the first embodiment
  • FIG. 2 is a diagram schematically illustrating the operation of the binary output unit 103 according to the first embodiment
  • FIG. 3 is a configuration diagram of a standard deviation estimation unit 106 according to Embodiment 1.
  • FIG. 1 is a configuration diagram of the control device 100 and the control target
  • the control target device 10 includes a roll-to-roll conveyance mechanism 11 that conveys and winds a conveyance material 12, a tension axis speed controller 21 that controls the rotation speed of the tension axis motor 13, and a speed axis motor 15. Are provided with a speed axis speed controller 22 for controlling the rotation speed and an adder 23 for performing addition.
  • the control target device 10 receives the tension axis speed addition value Vadd from the control device 100 and the reference speed command value Vr0 from the outside.
  • the inter-roll conveyance mechanism 11 has a mechanism for conveying the belt-like or linear conveyance material 12 between a plurality of rolls.
  • the conveying material 12 is exemplified by paper, resin, metal, or fiber.
  • the inter-roll transport mechanism 11 includes a tension shaft motor 13 having a drive mechanism, a tension shaft roll 14 having a rotation mechanism, a speed axis motor 15 having a drive mechanism, a speed shaft roll 16 having a rotation mechanism, and a detected tension value.
  • the inter-roll conveyance mechanism 11 winds the conveyance material 12 by driving the tension axis motor 13 and the speed axis motor 15 to rotate the tension axis roll 14 and the speed axis roll 16.
  • the slip between the tension shaft roll 14 and the conveying material 12 is assumed to be minute. It is assumed that the peripheral speed of the tension shaft roll 14 and the speed of the portion of the conveying material 12 in contact with the tension shaft roll 14 match or fall within a predetermined range.
  • the slip between the speed axis roll 16 and the conveying material 12 is assumed to be minute. It is assumed that the peripheral speed of the speed axis roll 16 and the speed of the portion of the conveying material 12 in contact with the speed axis roll 16 match or fall within a predetermined range.
  • the tension axis speed controller 21 controls the rotation speed of the tension axis motor 13 so that the speed at which the tension axis roll 14 conveys the conveying material 12 matches the input tension axis speed command value Vr1. Specifically, the tension axis speed controller 21 considers the diameter of the tension axis roll 14 and the reduction ratio, and in response to a command obtained by converting the tension axis speed command value Vr1 into the rotation speed of the tension axis motor 13, the tension axis motor Control is performed so that the rotational speeds of 13 coincide or fall within a predetermined range.
  • the speed axis speed controller 22 matches the input reference speed command value Vr0 so that the speed at which the speed axis roll 16 conveys the conveying material 12 matches or falls within a predetermined range. To control. Specifically, the speed axis speed controller 22 considers the diameter of the speed axis roll 16 and the reduction ratio, and in response to the command obtained by converting the reference speed command value Vr0 into the rotation speed of the speed axis motor 15, the speed axis motor. Control is performed so that the rotational speeds of 15 coincide or fall within a predetermined range.
  • the reference speed command value Vr0 input from the outside regulates the transport speed of the transport material 12, and can take various values depending on the transport conditions of the transport material 12.
  • the adder 23 adds the reference speed command value Vr0 and the tension axis speed addition value Vadd to calculate the tension axis speed command value Vr1, and outputs the tension axis speed command value Vr1.
  • the tension detector 20 outputs a tension detection value Yfb that is a value obtained by detecting the tension of the conveying material 12.
  • the tension detection value Yfb is a controlled variable and is a variable controlled so as to approach the command value as will be described later.
  • the tension shaft roll 14 is described as a configuration for winding the conveyance material 12, and the speed axis roll 16 is configured as a configuration for unwinding the conveyance material 12, but is not limited to these configurations.
  • the speed axis roll 16 may take up the conveying material 12, and the tension axis roll 14 may be configured to unwind the conveying material 12.
  • the tension shaft roll 14 and the speed shaft roll 16 may be intermediate shafts that perform only the feeding operation of the transporting material 12 without directly winding and unwinding the transporting material 12.
  • the control target device 10 adds the reference speed command value Vr0 and the tension axis speed addition value Vadd to generate the tension axis speed command value Vr1 of the tension axis roll 14. Therefore, the tension shaft roll 14 rotates faster than the speed shaft roll 16 by the tension shaft speed addition value Vadd. That is, the conveyance speed of the part of the conveying material 12 that contacts the tension axis roll 14 is faster than the conveyance speed of the part of the conveying material 12 that contacts the speed axis roll 16.
  • the conveying material 12 is tensioned by being pulled between the tension shaft roll 14 and the speed shaft roll 16.
  • the control target apparatus 10 changes the rotation speed of the tension axis roll 14, so that the tension generated in the conveying material 12 changes.
  • the inter-roll conveyance mechanism 11 measures the tension acting on the conveyance material 12 by the tension detector 20 and outputs a tension detection value Yfb. That is, the control target device 10 is configured to perform feedback control by being combined with the control device 100 that calculates the tension axis velocity addition value Vadd using the tension detection value Yfb.
  • the control device 100 includes a control calculation unit 101 that calculates an operation amount Uc, an adjustment execution command generation unit 102 that generates an adjustment execution command value SWat that is a command value indicating whether or not adjustment can be executed, and an operation amount Uc during adjustment.
  • a binary output unit 103 that calculates an adjustment addition value Uadd that is a value to be added, and a control gain calculation unit 104 that calculates a gain candidate value.
  • the control device 100 includes a control gain adjusting unit 105 that changes a gain value used in the control calculation unit 101, a standard deviation estimating unit 106 that calculates a standard deviation estimated value Sig that is an estimated value of the standard deviation, and a hysteresis width.
  • a hysteresis width calculation unit 107 that calculates a calculation value Hc, a subtracter 108 that performs subtraction, and an adder 109 that performs addition are provided.
  • the control device 100 receives a tension command value Yr from the outside, receives a tension detection value Yfb from the control target device 10, and outputs a tension axis speed addition value Vadd to the control target device 10.
  • the control calculation unit 101 receives a tension deviation value Ye that is a deviation between the tension command value Yr and the tension detection value Yfb, and an adjustment execution command value SWat.
  • the tension deviation value Ye is a control deviation.
  • the control calculation unit 101 performs proportional compensation obtained by multiplying the tension deviation value Ye by a proportional gain, which is one of the control gains, and tension.
  • the sum of the deviation value Ye and the integral compensation obtained by multiplying the integral gain, which is one of the control gains, is output as the manipulated variable Uc.
  • a period during which the adjustment execution command value SWat is on is referred to as an automatic adjustment period.
  • the control calculation unit 101 When the adjustment execution command value SWat switches from OFF to ON, the control calculation unit 101 holds the value of the operation amount Uc immediately before the adjustment execution command value SWat is turned ON, and the operation held during the automatic adjustment period. The value of the quantity Uc is output. The operation of maintaining the value of the operation amount Uc immediately before the adjustment execution command value SWat is turned on is realized by setting the proportional gain and the integral gain to 0 and holding the integral output. As a result, the control calculation unit 101 can maintain the immediately previous stable control state even in the automatic adjustment period, and stably shifts to the automatic adjustment period in which automatic adjustment is executed as will be described later. The control gain can be set to a highly accurate value.
  • the adjustment execution command generation unit 102 generates an adjustment execution command value SWat that is a signal indicating whether the adjustment execution command is on or off based on an instruction input by an external operation.
  • the adjustment execution command generation unit 102 changes the adjustment execution command value SWat from off to on by an external operation, outputs an on signal for a predetermined period, and then returns it to off.
  • the predetermined period is a period until a determination is made that the output of the binary output unit 103 to be described later has changed by a predetermined number of times or a predetermined number of times. Not limited to.
  • the binary output unit 103 performs an adjustment operation during the automatic adjustment period in which the adjustment execution command value SWat is on, and uses the tension deviation value Ye and the hysteresis width setting value Hs to set a large value of the added value amplitude D that is set in advance.
  • the adjustment added value Uadd which is a value having the same amplitude and for which positive / negative is determined by a method described later, is output. As will be described later, the binary output unit 103 outputs the adjustment addition value Uadd of 0 during the period when the adjustment execution command value SWat is off.
  • FIG. 2 schematically shows the operation of the binary output unit 103 that selects either + D or ⁇ D based on the tension deviation value Ye and the hysteresis width setting value Hs and outputs the adjustment added value Uadd.
  • FIG. 2 schematically shows the operation of the binary output unit 103 that selects either + D or ⁇ D based on the tension deviation value Ye and the hysteresis width setting value Hs and outputs the adjustment added value Uadd.
  • the binary output unit 103 selects one of the two values + D or -D based on the sign of the tension deviation value Ye. .
  • the binary output unit 103 outputs + D if the previous adjustment addition value Uadd is + D, and the previous adjustment addition value Uadd is ⁇ . If it is D, -D is selected.
  • the tension deviation value Ye has a hysteresis characteristic that is equal to the hysteresis width setting value Hs when determining the adjustment addition value Uadd.
  • FIG. 2 schematically shows the operation of the binary output unit 103 when one of the two values + D and ⁇ D is selected based on the tension deviation value Ye and the hysteresis width setting value Hs. ing.
  • the binary output unit 103 uses a signal obtained by applying a low-pass filter to the tension deviation value Ye instead of the tension deviation value Ye, and uses the tension deviation value Ye.
  • One of the two values + D and -D may be selected based on a signal obtained by applying a low-pass filter to the hysteresis width setting value Hs.
  • the operation of the binary output unit 103 is similar to a method called a limit cycle method used in temperature adjustment control, and when the adjustment execution command value SWat is turned on, the adjustment addition value Uadd output by the binary output unit 103;
  • the tension deviation value Ye oscillates.
  • the control gain calculation unit 104 receives the tension deviation value Ye and the adjustment execution command value SWat. In addition, the control gain calculation unit 104 measures the vibration period and amplitude of the tension deviation value Ye in the automatic adjustment period in which the adjustment execution command value SWat is on, and based on the measurement result, the control gain calculation unit 101 A proportional gain candidate value G1, which is a proportional gain candidate, and an integral gain candidate value G2, which is an integral gain candidate, are calculated.
  • the proportional gain candidate value G1 and the integral gain candidate value G2 are collectively referred to as a control gain candidate value.
  • control gain calculation unit 104 sets the value obtained by subtracting the hysteresis width setting value Hs from the amplitude of the tension deviation value Ye as the correction amplitude Ya, and multiplies the inverse of the correction amplitude Ya by a predetermined constant.
  • the proportional gain candidate value G1 is calculated.
  • the control gain calculation unit 104 multiplies the oscillation time by the ratio of the amplitude of the correction amplitude Ya and the tension deviation value Ye by a predetermined constant, and the integral gain candidate value G2 by multiplying the vibration time by the integral time constant of the proportional integral calculation. calculate.
  • the linearization gain of the input / output of the binary output unit 103 is calculated based on the description function method, and the proportional gain and the integral gain are calculated based on the Ziegler-Nichols limit sensitivity method. It is sufficient to use a method of determining With this method, the control gain calculation unit 104 can perform an accurate adjustment based on the characteristics of the conveying material 12 and the characteristics of the tension detector 20.
  • the control gain calculation unit 104 outputs the calculated proportional gain candidate value G1 and integral gain candidate value G2 when the adjustment execution command value SWat is turned off.
  • the control gain adjustment unit 105 receives the proportional gain candidate value G1 and the integral gain candidate value G2 calculated by the control gain calculation unit 104.
  • the control gain adjustment unit 105 changes the proportional gain and integral gain of the control calculation unit 101 to the calculated proportional gain candidate value G1 and integral gain candidate value G2.
  • the control gain adjustment unit 105 uses the proportional gain and integral gain of the control calculation unit 101 as the proportional gain candidate value G1 immediately after the proportional gain candidate value G1 and the integral gain candidate value G2 are input. Although it shall change to integral gain candidate value G2, it is not restricted to these structures. Even if the operator of the control device 100 confirms the proportional gain candidate value G1 and the integral gain candidate value G2, the control gain adjustment unit 105 executes the operation of changing the proportional gain and integral gain of the control calculation unit 101. Good.
  • the control gain adjustment unit 105 holds a plurality of sets of proportional gain candidate values G1 and integral gain candidate values G2, and the operator of the control device 100 selects one set of proportional gain candidates from the plurality of sets. After selecting the integral gain candidate value G2, the process of changing the proportional gain and integral gain of the control calculation unit 101 may be executed.
  • FIG. 3 is a specific configuration diagram of the standard deviation estimation unit 106.
  • the standard deviation estimation unit 106 includes a high-pass filter unit 106a that calculates a low-frequency component removal signal yh from which a low-frequency component of the tension detection value Yfb is removed, and an output signal y 106b that is a square value of the input signal yh. And a low-pass filter unit 106c for calculating the output signal y 106c by applying a low-pass filter. Further, the standard deviation estimation unit 106 calculates a standard root estimation value Sig based on the square root calculation unit 106d that calculates the output signal y 106d that is the square root of the input signal y 106c , and the adjustment execution command value SWat. A selector unit 106e for outputting and a delay unit 106f for holding the standard deviation estimated value Sig are provided.
  • the standard deviation estimator 106 receives the adjustment execution command value SWat and the tension detection value Yfb, and during the period when the adjustment execution command value SWat is off, the standard deviation estimation value is calculated based on the tension detection value Yfb by a method described later. Sig is calculated and a standard deviation estimated value Sig is output.
  • the standard deviation estimator 106 operates so as to output a previous standard deviation estimated value Sig ⁇ that is a standard deviation estimated value one cycle before, which will be described later, during an automatic adjustment period in which the adjustment execution command value SWat is on. Therefore, the standard deviation estimated value Sig output from the standard deviation estimating unit 106 during the automatic adjustment period becomes the previous standard deviation estimated value Sig ⁇ .
  • the high-pass filter unit 106a receives the tension detection value Yfb, operates a high-pass filter on the tension detection value Yfb, and outputs a low-frequency component removal signal yh from which the low-frequency component of the tension detection value Yfb is removed. Calculate and output the low frequency component removal signal yh.
  • the square value calculation unit 106b receives the low frequency component removal signal yh, calculates an output signal y 106b obtained by squaring the value of the low frequency component removal signal yh, and outputs the output signal y 106b .
  • the low-pass filter unit 106c receives the output signal y 106b, applies a low-pass filter to the output signal y 106b , and calculates an output signal y 106c from which the high-frequency component of the output signal y 106b is removed.
  • the output signal y 106c is output.
  • the square root calculation unit 106d receives the output signal y 106c , calculates an output signal y 106d obtained by calculating the square root of the output signal y 106c , and outputs the output signal y 106d .
  • the selector unit 106e is the previous standard deviation estimate Sig - are inputted and an output signal y 106d, based on the adjustment execution command value SWAT, last time when adjustment execution command value SWAT is on the standard deviation estimation value Sig - Select When the adjustment execution command value SWat is off, the output signal y 106d is selected, and the selected signal is output as the standard deviation estimated value Sig.
  • the delay unit 106f holds the standard deviation estimated value Sig for only one period of the operating cycle of the standard deviation estimating unit 106, and sets the held value as the previous standard deviation estimated value Sig ⁇ to the operating cycle of the standard deviation estimating unit 106. Is output after one cycle.
  • the hysteresis width calculation unit 107 receives the standard deviation estimated value Sig, and calculates the hysteresis width calculated value Hc based on the standard deviation estimated value Sig. Specifically, the hysteresis width calculation unit 107 calculates the hysteresis width calculation value Hc by multiplying the standard deviation estimated value Sig by a predetermined coefficient Kh.
  • the coefficient Kh is a constant and is set to a value between 1 and 60.
  • the hysteresis width calculation unit 107 changes the hysteresis width set value Hs of the binary output unit 103 to the hysteresis width calculation value Hc every operation cycle.
  • the subtractor 108 receives the tension command value Yr and the tension detection value Yfb, calculates the tension deviation value Ye from the difference between the tension command value Yr and the tension detection value Yfb, and outputs the tension deviation value Ye.
  • the adder 109 receives the operation amount Uc and the adjustment addition value Uadd, adds the operation amount Uc and the adjustment addition value Uadd, calculates the tension axis velocity addition value Vadd, and calculates the tension axis velocity addition value Vadd. Output.
  • the control calculation unit 101 calculates the operation amount Uc so that the tension deviation value Ye becomes 0 during a period in which the adjustment execution command value SWat output from the adjustment execution command generation unit 102 is off.
  • the adjustment addition value Uadd which is the output of the binary output unit 103, is 0, and the tension axis velocity addition value Vadd matches the value of the operation amount Uc. That is, the operation of outputting the tension axis velocity addition value Vadd by the control device 100 is an operation of feedback control by PI control.
  • the standard deviation estimation unit 106 outputs a standard deviation estimated value Sig that is an estimated value of the standard deviation of the low frequency component removal signal yh obtained by removing the low frequency component of the tension detection value Yfb during the period when the adjustment execution command value SWat is off. To do.
  • the hysteresis width calculation unit 107 calculates a hysteresis width calculation value Hc based on the standard deviation estimated value Sig calculated by the standard deviation estimation unit 106, and calculates the hysteresis width set value Hs of the binary output unit 103 as a hysteresis width calculation. Change to value Hc.
  • the adjustment execution command generation unit 102 changes the output of the adjustment execution command value SWat from off to on.
  • the control calculation unit 101 holds the output of the integration and outputs a certain operation amount Uc.
  • the standard deviation estimation unit 106 switches to an operation of substituting the previous standard deviation estimated value Sig ⁇ into the standard deviation estimated value Sig, and operates so that the previous standard deviation estimated value Sig ⁇ is output. Therefore, when the adjustment execution command value SWat is changed from OFF to ON, the standard deviation estimation value Sig output from the standard deviation estimation unit 106, the previous standard deviation estimation value Sig - becomes.
  • the hysteresis width calculation value Hc calculated by the hysteresis width calculation unit 107 is a constant value
  • the hysteresis width setting value Hs of the binary output unit 103 is a constant value.
  • the binary output unit 103 alternately selects + D or ⁇ D based on the tension deviation value Ye and the hysteresis width setting value Hs, and outputs the adjustment added value Uadd.
  • the tension axis speed addition value Vadd and the tension deviation value Ye are obtained when the hysteresis output value Yfb includes the hysteresis width setting value Hs when the + D or -D value of the adjustment addition value Uadd of the binary output unit 103 is selected. If the influence can be appropriately removed, limit cycle vibration having a constant or constant period is generated.
  • control gain calculation unit 104 calculates the proportional gain candidate value G1 and the integral gain candidate value G2 based on the vibration amplitude and vibration period of the tension deviation value Ye during the period when the adjustment execution command value SWat is on.
  • the adjustment execution command generation unit 102 changes the adjustment execution command value SWat from on to off after a predetermined automatic adjustment period has elapsed since the adjustment execution command value SWat was changed from off to on.
  • the binary output unit 103 holds the value of the adjustment addition value Uadd at 0 when the adjustment execution command generation unit 102 changes the adjustment execution command value SWat from on to off.
  • the control gain calculation unit 104 outputs the proportional gain candidate value G1 and the integral gain candidate value G2 calculated immediately before the adjustment execution command value SWat is changed from on to off.
  • the control gain adjustment unit 105 changes the proportional gain value and integral gain value of the control calculation unit 101 to the input proportional gain candidate value G1 and integral gain candidate value G2.
  • the standard deviation estimation unit 106 operates so that the output signal y 106d calculated by the square root calculation unit 106d is output. Therefore, the standard deviation estimated value Sig output from the standard deviation estimating unit 106 is the output signal y 106d .
  • the control calculation unit 101 starts calculating the manipulated variable Uc based on the proportional gain and the integral gain changed by the control gain adjustment unit 105.
  • the control device 100 sets the hysteresis width setting value Hs of the binary output unit 103 to an appropriate value, and sets the adjustment addition value Uadd based on the tension deviation value Ye and the hysteresis width setting value Hs.
  • the influence of noise included in the detected tension value Yfb is reduced.
  • the control device 100 can generate a limit cycle having a constant or constant period, and can adjust the control gain with high accuracy.
  • the control device 100 substitutes the calculated hysteresis width calculation value Hc for the hysteresis width setting value Hs.
  • the hysteresis width calculation value Hc is calculated based on the standard deviation estimated value Sig.
  • the standard deviation estimated value Sig is a good estimated value of the standard deviation of the noise signal included in the tension detection value Yfb by the standard deviation estimating unit 106 described later. That is, since the control device 100 estimates the standard deviation value of the noise signal included in the detected tension value Yfb based on the standard deviation estimated value Sig, it is possible to estimate the distribution of the noise signal generated during the automatic adjustment period.
  • the hysteresis width calculation value Hc having a magnitude that exceeds the amplitude of the noise signal can be calculated with high probability.
  • the high-pass filter unit 106a calculates a low-frequency component removal signal yh obtained by applying a high-pass filter to the tension detection value Yfb during the period when the adjustment execution command value SWat is off.
  • the tension detection value Yfb is a low-frequency vibration component due to the influence of a disturbance and a tension command value. It has an offset error with respect to Yr.
  • the low frequency vibration component and the offset error are caused by the control band being low and the control performance being not appropriate, and are observed during the operation of the limit cycle method in which the tension detection value Yfb is vibrated at a high frequency near the limit frequency. This phenomenon is not possible. That is, the low-frequency vibration component and the offset error caused by the low control band do not affect the determination of the adjustment addition value Uadd in the limit cycle method. There is no need to consider.
  • the vibration amplitude of the low frequency component of the tension detection value Yfb may be larger than the amplitude of the noise, and it is standard without distinguishing the vibration of the low frequency component of the tension detection value Yfb from the noise signal. If the deviation estimated value Sig is calculated, the calculation result of the standard deviation estimated value Sig greatly exceeds the true value of the standard deviation of the noise signal. Since the hysteresis width setting value Hs that is larger than necessary is set and the vibration amplitude of the tension detection value Yfb is excessively amplified, the vibration component and the offset error of the low frequency component are the standard deviation estimated value. It should be removed when calculating Sig.
  • the high-pass filter unit 106a causes the high-pass filter to act on the tension detection value Yfb, the low-frequency vibration component and the offset error of the tension detection value Yfb are removed, and noise composed of the high-frequency component of the tension detection value Yfb. Can be extracted by the low frequency component removal signal yh.
  • the standard deviation estimation unit 106 including the square value calculation unit 106b, the low-pass filter unit 106c, and the square root calculation unit 106d will be described.
  • a time series signal x having a certain normal distribution is taken as an example.
  • the average value of the time series signal x is ⁇ x
  • the standard deviation is ⁇ x .
  • xs (i) indicates the i-th sample extracted from the time series signal x.
  • Xs (n) represents a sample at the end time in the time-series signal x.
  • T N ⁇ dt.
  • xlpf (i) the i-th sample of a signal obtained by applying a first-order lag low-pass filter having a time constant ⁇ larger than dt to the time series signal x is assumed to be xlpf (i).
  • xlpf (n) at the end time is calculated by the following equation (3).
  • N is the number of samples of the time series signal x used for the calculation of xlpf (n).
  • Equation is approximated using the fact that the time constant ⁇ is larger than dt and that N is larger than 1. Further, o (dt 2 / ⁇ 2 ) is a summary of terms having a magnitude of the order of dt 2 / ⁇ 2 or less, and can be approximated to 0.
  • the standard deviation estimation unit 106 calculates an estimated value of the standard deviation of the low frequency component removal signal yh by a method using a low-pass filter as in the above example.
  • the low frequency component removal signal yh is a signal from which the low frequency component of the tension detection value Yfb is removed
  • the average value of the low frequency component removal signal yh can be approximated to zero. That is, the estimated value of the standard deviation of the low frequency component removal signal yh can be calculated based on Expression (7) that is an arithmetic expression of the estimated value of the standard deviation when the average value is 0.
  • the standard deviation estimation unit 106 can estimate the standard deviation of the noise signal included in the detected tension value Yfb.
  • the above-described time series signal x is taken as an example.
  • the average value ⁇ x of the time series signal x is considered to be zero. Since the time series signal x has a normal distribution, the sample xs randomly extracted from the time series signal x is a value that satisfies the following expression (8) with a probability of about 68.26889492%, and about 99 It is a value that satisfies the following equation (9) with a probability of 9999998%.
  • the upper and lower limits of the range in which the sample is included with a high probability is a high probability that a randomly sampled xs is approximately 68.2688992% to approximately 99.99999998%. Is included in the range. Therefore, it can be said that a value obtained by multiplying the standard deviation by a coefficient of 1 to 6 is a good estimated value of the amplitude that the time-series signal x can take within a certain period.
  • the value obtained by multiplying the standard deviation estimated value Sig, which is the estimated value of the noise signal included in the tension detection value Yfb, by a coefficient of 1 to 6 is the amplitude of the noise signal within a certain period. It can be said that this is a good estimate.
  • the hysteresis calculation unit 107 sets the magnitude of the hysteresis to 1 to 10 times the amplitude of the noise included in the tension detection value Yfb, so that the influence of the noise is removed and the hysteresis An excessively large size can be avoided.
  • the hysteresis calculation unit 107 calculates the hysteresis width calculation value Hc by multiplying the standard deviation estimated value Sig, which is an estimated value of the standard deviation of the noise signal included in the tension detection value Yfb, by a coefficient Kh of 1 to 60. To do.
  • the coefficient Kh of 1 or more and 60 or less was intended to set a value obtained by multiplying the standard deviation by a coefficient of 1 or more and 6 or less and further multiply by a coefficient of 1 or more and 10 or less as the hysteresis calculation value Hc. Is.
  • the magnitude of the hysteresis width calculation value Hc is calculated to such an extent that the influence of noise can be removed and the vibration amplitude of the controlled variable is not increased excessively.
  • the control device 100 can remove the influence of noise and can appropriately set the magnitude of hysteresis so as not to excessively increase the vibration amplitude of the controlled variable.
  • control apparatus 100 does not need to be provided with the function to display the control amount, and the function to set the magnitude
  • the control device 100 sequentially calculates an appropriate hysteresis magnitude by the functions of the standard deviation estimation unit 106 and the hysteresis width calculation unit 107. The For this reason, the control apparatus 100 is appropriate even if there is a change in the magnitude of the control amount and a change in the magnitude of noise due to a change in the operating state of the electric machine and the electronic device installed around the control target apparatus 10 A large hysteresis value can be set.
  • control device 100 can set an appropriate magnitude of hysteresis and appropriately determine the adjustment addition value during the automatic adjustment period, so that hunting due to the influence of noise included in the tension detection value Yfb can be suppressed. .
  • control device 100 can generate a limit cycle having a constant or constant period, and can accurately calculate the value of the control gain.
  • the high-pass filter unit 106a has been described as having a configuration in which the high-pass filter is applied to the tension detection value Yfb to remove the low-frequency component of the tension detection value Yfb. I can't.
  • the high-pass filter unit 106a removes the difference between the tension detection value Yfb and the constant when the tension detection value Yfb shows a constant value with the lapse of time, excluding the signal of the high frequency component. You may comprise by the subtractor calculated with the signal yh. Even in this configuration, the high-pass filter unit 106a can calculate the low-frequency component removal signal yh from which the low-frequency component of the tension detection value Yfb is removed, and it is clear that the same effect can be obtained. The same applies to the following embodiments.
  • control device 100 in which the low-pass filter is configured by a first-order lag filter has been described.
  • the control device 100 only needs to have a characteristic of removing high frequency components of the input signal, and has a different pole arrangement such as a high-order lag filter, a Butterworth filter, or a Chebyshev filter in which first-order lag filters are arranged in series. Even if a filter is used, the same effect can be obtained. The same applies to the following embodiments.
  • the hysteresis width calculation unit 107 changes the hysteresis width setting value Hs of the binary output unit 103 to the hysteresis width calculation value Hc for each operation cycle, the present invention is not limited to these configurations.
  • the hysteresis width calculation unit 107 may change the hysteresis width setting value Hs of the binary output unit 103 to the hysteresis width calculation value Hc every time that is an integral multiple of the operation cycle.
  • the hysteresis width calculation unit 107 has a function that allows the operator to specify the timing for changing the hysteresis width setting value Hs during the operation of the control device 100, and the hysteresis width setting value at the timing specified by the operator. Hs may be changed to the hysteresis width calculation value Hc. The same applies to the following embodiments.
  • control target device 10 is described as a device including an inter-roll conveyance mechanism, but is not limited to an apparatus including an inter-roll conveyance mechanism.
  • the control target device 10 has a mechanism that can change the control amount by inputting an operation amount from the outside, and a mechanism that detects and outputs the control amount, and can be stabilized by feedback control.
  • An apparatus having another configuration may be used. The same applies to the following embodiments.
  • FIG. A second embodiment of the control device according to the present invention will be described.
  • 4 is a configuration diagram of the control device 200 and the control target device 10 according to the second embodiment
  • FIG. 5 is a configuration diagram of the standard deviation estimation unit 116.
  • the standard deviation estimation unit 116 is obtained by changing the standard deviation estimation unit 106 of the control device 100 described above. Note that components having functions similar to those of the first embodiment are denoted by the same reference numerals as those of the first embodiment, and detailed description thereof is omitted.
  • the control device 200 includes a control calculation unit 101 that calculates an operation amount Uc, an adjustment execution command generation unit 102 that generates an adjustment execution command value SWat, a binary output unit 103 that calculates an adjustment addition value Uadd, and gain candidates And a control gain calculation unit 104 that calculates a value.
  • the control device 200 also includes a control gain adjustment unit 105 that changes the gain value used in the control calculation unit 101, a standard deviation estimation unit 116 that calculates a standard deviation estimation value Sig, and a hysteresis that calculates a hysteresis width calculation value Hc.
  • a width calculation unit 107, a subtracter 108 that performs subtraction, and an adder 109 that performs addition are provided.
  • the control device 200 receives the tension command value Yr from the outside, receives the tension detection value Yfb from the control target device 10, and outputs the tension axis speed addition value Vadd to the control target device 10.
  • the standard deviation estimation unit 116 includes a high-pass filter unit 106a that calculates the low frequency component removal signal yh, and an absolute value calculation unit that calculates the absolute value of the value of the low frequency component removal signal yh and calculates the output signal y 116b. 116b, and a low-pass filter unit 116c that calculates the output signal y 116c by operating a low-pass filter.
  • the standard deviation estimation unit 116 also includes a conversion gain unit 116d that calculates the output signal y 116d based on the output signal y 116c , a selector unit 116e that outputs the standard deviation estimated value Sig based on the adjustment execution command value SWat, A delay unit 106f that holds the standard deviation estimated value Sig.
  • the standard deviation estimation unit 116 receives the adjustment execution command value SWat and the tension detection value Yfb, calculates a standard deviation estimation value Sig based on the tension detection value Yfb during the period when the adjustment execution command value SWat is off, The standard deviation estimated value Sig is output.
  • the standard deviation estimation unit 116 outputs the previous standard deviation estimated value Sig ⁇ during the automatic adjustment period in which the adjustment execution command value SWat is on. Therefore, the standard deviation estimated value Sig output from the standard deviation estimating unit 116 during the automatic adjustment period becomes the previous standard deviation estimated value Sig ⁇ .
  • the absolute value calculator 116b receives the low frequency component removal signal yh, calculates an output signal y 116b obtained by calculating an absolute value with respect to the value of the low frequency component removal signal yh, and outputs an output signal y 116b . To do.
  • the low-pass filter unit 116c receives the output signal y 116b, applies a low-pass filter to the output signal y 116b , calculates the output signal y 116c from which the high-frequency component of the output signal y 116b is removed, and outputs the output signal y 116c.
  • the signal y 116c is output.
  • Conversion gain circuit 116d is supplied with the output signal y 116c, ⁇ ( ⁇ / 2 ) multiplied by the calculated output signal y 116d for the output signal y 116c, and outputs an output signal y 116d. Note that ⁇ ( ⁇ / 2) means that ( ⁇ / 2) is a root sign.
  • the selector 116e receives the previous standard deviation estimated value Sig ⁇ and the output signal y 116d .
  • the selector unit 116e based on the adjustment execution command value SWAT, last time when adjustment execution command value SWAT is on the standard deviation estimation value Sig - select, also, adjustment execution command value SWAT is an output signal y 116d is in the off The selected one is output as the standard deviation estimated value Sig.
  • the standard deviation estimation unit 116 is the standard deviation estimated value Sig that is an estimated value of the standard deviation of the low frequency component removal signal yh from which the low frequency component of the tension detection value Yfb is removed during the period when the adjustment execution command value SWat is off. Is output.
  • control device 200 sets the hysteresis width setting value Hs of the binary output unit 103 to an appropriate value, and determines the adjustment addition value Uadd based on the tension deviation value Ye and the hysteresis width setting value Hs.
  • the influence of noise included in the tension detection value Yfb is reduced.
  • the control device 200 can generate a limit cycle having a constant or constant period, and can adjust the control gain with high accuracy.
  • the control device 200 substitutes the calculated hysteresis width calculation value Hc for the hysteresis width setting value Hs.
  • the hysteresis width calculation value Hc is calculated based on the standard deviation estimated value Sig.
  • the standard deviation estimated value Sig is a good estimated value of the standard deviation of the noise signal included in the tension detection value Yfb by the standard deviation estimating unit 116 described later. That is, since the control device 200 estimates the standard deviation value of the noise signal included in the tension detection value Yfb based on the standard deviation estimated value Sig, it is possible to estimate the distribution of the noise signal generated during the automatic adjustment period.
  • the hysteresis width calculation value Hc having a magnitude that exceeds the amplitude of the noise signal can be calculated with high probability.
  • the control device 200 and the control device 100 described above differ only in the configuration of the standard deviation estimation unit.
  • operations and effects when the standard deviation estimation unit 116 calculates the standard deviation estimated value Sig will be described.
  • a time series signal x having a certain normal distribution is taken as an example.
  • the average value of the time series signal x is ⁇ x
  • the standard deviation is ⁇ x
  • a signal obtained by taking the absolute value of the time series signal x is defined as a signal ax
  • an average value of the signal ax is defined as ⁇ ax .
  • the value obtained by multiplying ⁇ a ( ⁇ / 2) relative to the average value mu ax signal ax is a standard deviation of the time series signal x.
  • ⁇ ( ⁇ / 2) means that ( ⁇ / 2) is a root sign.
  • the expressions (1) and (4) described in the first embodiment are similarly established for the signal ax.
  • a sample is extracted from the signal ax at a time interval dt, and the i-th sample is axs (i), and axs (n) is the sample of the signal ax at the end time.
  • the i-th sample of a signal obtained by applying a first-order lag low-pass filter with a time constant ⁇ larger than dt to the signal ax is defined as axlpf (i).
  • the following expressions (11) and (12) hold.
  • Equation (11) when the number of samples N is large and the time constant ⁇ of the low-pass filter is larger than the sample time interval dt, axlpf (n) and ⁇ ax are It can be seen that axlpf (n) is a good estimate of ⁇ ax because it matches or falls within a predetermined range.
  • the absolute value calculation unit 116b, the low-pass filter unit 116c, and the conversion gain unit 116d perform the same calculation as in the above-described example. That is, the output signal y 116c is an estimated value with a good average value of the output signal y 116b . Further, the output signal y 116d is an estimated value with a good standard deviation of the low frequency component removal signal yh.
  • control device 200 can set an appropriate magnitude of hysteresis, and appropriately determine the adjustment addition value during the automatic adjustment period, so that hunting due to the influence of noise included in the tension detection value Yfb can be suppressed. .
  • control device 200 can generate a limit cycle having a constant or constant period, and can accurately calculate the value of the control gain.
  • FIG. 6 is a configuration diagram of the control device 300 and the control target device 10 according to the third embodiment
  • FIG. 7 is a configuration diagram of the standard deviation estimation unit 126.
  • the standard deviation estimation unit 126 is obtained by changing the standard deviation estimation unit 116 of the control device 200 described above. Note that components having the same functions as those in the first embodiment or the second embodiment are denoted by the same reference numerals as those in the first or second embodiment, and detailed description thereof is omitted.
  • the control apparatus 300 includes a control calculation unit 101 that calculates an operation amount Uc, an adjustment execution command generation unit 102 that generates an adjustment execution command value SWat, a binary output unit 103 that calculates an adjustment addition value Uadd, and gain candidates And a control gain calculation unit 104 that calculates a value.
  • the control device 300 includes a control gain adjustment unit 105 that changes a gain value used in the control calculation unit 101, a standard deviation estimation unit 126 that calculates a standard deviation estimation value Sig, and a hysteresis that calculates a hysteresis width calculation value Hc.
  • a width calculation unit 107, a subtracter 108 that performs subtraction, and an adder 109 that performs addition are provided.
  • the control device 300 receives the tension command value Yr from the outside, receives the tension detection value Yfb from the control target device 10, and outputs the tension axis speed addition value Vadd to the control target device 10.
  • the standard deviation estimation unit 126 includes a high-pass filter unit 106a that calculates the low frequency component removal signal yh, and an absolute value calculation unit that calculates the absolute value of the value of the low frequency component removal signal yh and calculates the output signal y 116b. 116b, a first low-pass filter unit 126c that calculates an output signal y 126c by applying a low-pass filter, and a first square value calculation unit 126d that calculates an output signal y 126d by square value calculation A second square value calculation unit 126e that calculates the output signal y 126e by square value calculation, and a second low-pass filter unit 126f that calculates the output signal y 126f by applying a low-pass filter. Is provided.
  • the standard deviation estimation unit 126 calculates a subtracter 126g performing subtraction, and square root calculator 126h for calculating an output signal y 126h by the square root calculation, the output signal y 126i on the basis of the output signal y 126h
  • a conversion gain unit 126i that outputs the standard deviation estimated value Sig based on the adjustment execution command value SWat, and a delay unit 106f that holds the standard deviation estimated value Sig.
  • the standard deviation estimating unit 126 calculates the standard deviation estimated value Sig based on the tension detection value Yfb in a period in which the adjustment execution command value SWat and the tension detection value Yfb are input and the adjustment execution command value SWat is off, The standard deviation estimated value Sig is output.
  • the standard deviation estimation unit 126 operates so that the previous standard deviation estimated value Sig ⁇ is output in the automatic adjustment period in which the adjustment execution command value SWat is on. Therefore, the standard deviation estimated value Sig output from the standard deviation estimating unit 126 during the automatic adjustment period becomes the previous standard deviation estimated value Sig ⁇ .
  • the output signal y 116 b is input, by applying a low pass filter to the output signal y 116 b, the output signal y 116 b the high-frequency component output signal to remove y 126c of And the output signal y 126c is output.
  • the first square value calculator 126d receives the output signal y 126c , calculates the output signal y 126d obtained by squaring the output signal y 126c , and outputs the output signal y 126d .
  • the second square value calculator 126e receives the output signal y 116b , calculates the output signal y 126e obtained by squaring the output signal y 116b , and outputs the output signal y 126e .
  • Second low-pass filter unit 126f the output signal y 126 e is input, by applying a low pass filter to the output signal y 126 e, the output signal y output signal to remove high-frequency components of 126 e y 126f And the output signal y 126f is output.
  • Subtractor 126g computes the difference between the output signal y 126d and the output signal y 126 f to calculate the output signal y 126g, and outputs an output signal y 126g.
  • Square root calculation unit 126h the output signal y 126 g is inputted, calculates the output signal y 126h obtained by calculating the square root of the output signal y 126 g, and outputs an output signal y 126h.
  • Conversion gain unit 126i an output signal y 126h is input, ⁇ ( ⁇ / ( ⁇ - 2)) multiplied by the calculated output signal y 126i to the output signal y 126h, and outputs an output signal y 126i.
  • ⁇ ( ⁇ / ( ⁇ -2)) means that ( ⁇ / ( ⁇ -2)) is a root sign.
  • the selector unit 126j may last standard deviation estimate Sig - and the output signal y 126i is inputted.
  • the selector unit 126j based on the adjustment execution command value SWAT, last time when adjustment execution command value SWAT is on the standard deviation estimation value Sig - select, also, adjustment execution command value SWAT is an output signal y 126i is in the off The selected one is output as the standard deviation estimated value Sig.
  • the standard deviation estimation unit 126 estimates the standard deviation estimated value Sig that is an estimated value of the standard deviation of the low frequency component removal signal yh obtained by removing the low frequency component of the tension detection value Yfb during the period when the adjustment execution command value SWat is off. Is output.
  • the control device 300 sets the hysteresis width setting value Hs of the binary output unit 103 to an appropriate value, and determines the adjustment addition value Uadd based on the tension deviation value Ye and the hysteresis width setting value Hs. The influence of noise included in the value Yfb is reduced. As a result, the controller 300 can generate a limit cycle having a constant or constant period, and can adjust the control gain with high accuracy.
  • the control device 300 substitutes the calculated hysteresis width calculation value Hc for the hysteresis width setting value Hs.
  • the hysteresis width calculation value Hc is calculated based on the standard deviation estimated value Sig.
  • the standard deviation estimated value Sig is a good estimated value of the standard deviation of the noise signal included in the tension detection value Yfb by the standard deviation estimating unit 126 described later. That is, since the control device 300 estimates the standard deviation value of the noise signal included in the tension detection value Yfb based on the standard deviation estimated value Sig, it is possible to estimate the distribution of the noise signal generated during the automatic adjustment period.
  • the hysteresis width calculation value Hc having a magnitude that exceeds the amplitude of the noise signal can be calculated with high probability.
  • the control device 300 and the control device 100 differ only in the configuration of the standard deviation estimation unit.
  • operations and effects when the standard deviation estimation unit 126 calculates the standard deviation estimated value Sig will be described.
  • a time series signal x having a certain normal distribution is taken as an example.
  • the average value of the time series signal x is ⁇ x
  • the standard deviation is ⁇ x
  • a signal obtained by taking the absolute value of the time series signal x is set as a signal ax
  • an average value of the signal ax is set as ⁇ ax
  • a standard deviation is set as ⁇ ax .
  • the value obtained by multiplying the standard deviation ⁇ ax of the signal ax by ⁇ ( ⁇ / ( ⁇ 2)) is the standard deviation of the time series signal x.
  • ⁇ ( ⁇ / ( ⁇ -2)) means that ( ⁇ / ( ⁇ -2)) is a root sign.
  • a sample is extracted from the signal ax at a time interval dt, the i-th sample is axs (i), and axs (n) is a sample of the signal ax at the end time.
  • the i-th sample of a signal obtained by applying a first-order-lag low-pass filter with a time constant ⁇ larger than dt to the signal ax is defined as axlpf (i).
  • the standard deviation ⁇ ax of the signal ax is calculated by the following equation (14).
  • ax2lpf (n) ax2lpf (n) is Is calculated by the following equation (15).
  • the square root calculation unit 126h and the conversion gain unit 126i perform the same calculation as in the above-described example. That is, the output signal y 126h is an estimated value with a good standard deviation of the output signal y 116b .
  • the output signal y 126i is an estimated value with a good standard deviation of the low frequency component removal signal yh.
  • control device 300 can set an appropriately large hysteresis and appropriately determine the adjustment addition value during the automatic adjustment period, so that hunting due to the influence of noise included in the tension detection value Yfb can be suppressed. .
  • the control device 300 can generate a limit cycle having a constant or constant period, and can accurately calculate the value of the control gain.
  • FIG. Embodiment 4 of the control device according to the present invention will be described.
  • FIG. 8 is a configuration diagram of the control device 400 and the control target device 10 according to the fourth embodiment
  • FIG. 9 is a configuration diagram of the standard deviation estimation unit 136.
  • the standard deviation estimation unit 136 is a modification of the standard deviation estimation unit 106 of the control device 100 described above. Note that components having functions similar to those of the first embodiment are denoted by the same reference numerals as those of the first embodiment, and detailed description thereof is omitted.
  • the control device 400 includes a control calculation unit 101 that calculates an operation amount Uc, an adjustment execution command generation unit 102 that generates an adjustment execution command value SWat, a binary output unit 103 that calculates an adjustment addition value Uadd, and gain candidates And a control gain calculation unit 104 that calculates a value.
  • the control device 400 also includes a control gain adjustment unit 105 that changes the gain value used in the control calculation unit 101, a standard deviation estimation unit 136 that calculates a standard deviation estimation value Sig, and a hysteresis that calculates a hysteresis width calculation value Hc.
  • a width calculation unit 107, a subtracter 108 that performs subtraction, and an adder 109 that performs addition are provided.
  • the control device 400 receives the tension command value Yr from the outside, receives the tension detection value Yfb from the control target device 10, and outputs the tension axis speed addition value Vadd to the control target device 10.
  • the standard deviation estimation unit 136 includes a high-pass filter unit 106a that calculates the low-frequency component removal signal yh, a data holding unit 136b that holds the low-frequency component removal signal yh, and a standard deviation calculation unit 136c that calculates the standard deviation.
  • the selector 136d that outputs the standard deviation estimated value Sig based on the adjustment execution command value SWat and the delay unit 106f that holds the standard deviation estimated value Sig are provided.
  • the standard deviation estimator 136 calculates the standard deviation estimated value Sig based on the tension detection value Yfb during the period when the adjustment execution command value SWat and the tension detection value Yfb are input and the adjustment execution command value SWat is off.
  • the standard deviation estimated value Sig is output.
  • the standard deviation estimation unit 136 operates so that the previous standard deviation estimated value Sig ⁇ is output during the automatic adjustment period in which the adjustment execution command value SWat is on. Therefore, the standard deviation estimated value Sig output from the standard deviation estimating unit 136 during the automatic adjustment period becomes the previous standard deviation estimated value Sig ⁇ .
  • the data holding unit 136b receives the low frequency component removal signal yh, holds the value of the low frequency component removal signal yh with data associated with time, and outputs the latest m pieces of data among the held data as the output signal y. It outputs as 136b .
  • m is an integer of 1 or more.
  • the standard deviation calculator 136c receives the output signal y 136b , calculates the standard deviation of the output signal y 136b , and outputs the calculated output signal y 136c .
  • the output signal y 136b is m low-frequency component removal signals yh associated with time, and is a low-frequency component removal signal yh in a predetermined period, the output signal y 136b will be described below.
  • the standard deviation is referred to as a standard deviation for a predetermined period.
  • the selector unit 136d is previous standard deviation estimate Sig - are inputted and an output signal y 136c, based on the adjustment execution command value SWAT, last time when adjustment execution command value SWAT is on the standard deviation estimation value Sig - Select When the adjustment execution command value SWat is off, the output signal y 136c is selected, and the selected signal is output as the standard deviation estimated value Sig.
  • the standard deviation estimation unit 136 estimates the standard deviation estimated value Sig that is an estimated value of the standard deviation of the low frequency component removal signal yh from which the low frequency component of the tension detection value Yfb is removed during the period when the adjustment execution command value SWat is off. Is output.
  • control device 400 sets the hysteresis width setting value Hs of the binary output unit 103 to an appropriate value, and determines the adjustment addition value Uadd based on the tension deviation value Ye and the hysteresis width setting value Hs.
  • the influence of noise included in the tension detection value Yfb is reduced.
  • the control apparatus 400 can generate a limit cycle having a constant or constant period, and can adjust the control gain with high accuracy.
  • the control device 400 substitutes the calculated hysteresis width calculation value Hc for the hysteresis width setting value Hs.
  • the hysteresis width calculation value Hc is calculated based on the standard deviation estimated value Sig.
  • the standard deviation estimated value Sig is a good estimated value of the standard deviation of the noise signal included in the tension detection value Yfb by the standard deviation estimating unit 136 described later. That is, since the control device 400 estimates the standard deviation value of the noise signal included in the tension detection value Yfb based on the standard deviation estimated value Sig, it is possible to estimate the distribution of the noise signal generated during the automatic adjustment period.
  • the hysteresis width calculation value Hc having a magnitude that exceeds the amplitude of the noise signal can be calculated with high probability.
  • the control device 400 and the control device 100 differ only in the configuration of the standard deviation estimation unit.
  • operations and effects when the standard deviation estimation unit 136 calculates the standard deviation estimated value Sig will be described.
  • a time series signal x having a certain normal distribution is taken as an example.
  • the standard deviation of the time series signal x is ⁇ x .
  • the standard deviation ⁇ x is calculated when the number of samples xs is increased to infinity.
  • ⁇ xM and ⁇ xM are defined by Equation (17) and Equation (18) using a certain number M.
  • M is an integer of 1 or more.
  • Equation (2) Comparing Equation (2) and Equation (18), if M is a large number, ⁇ xM calculated by M samples xs and standard deviation ⁇ x are equal or within a predetermined range. Therefore , it can be said that ⁇ xM is a good estimate of the standard deviation ⁇ x .
  • ⁇ xM is the standard deviation of the time series signal x in a predetermined period, it is the above-mentioned specified period standard deviation.
  • the standard deviation calculation unit 136c performs the same calculation as in the above-described example. become. That is, it can be said that the output signal y 136c is a good estimate of the standard deviation of the low frequency component removal signal yh.
  • control device 400 can set an appropriate magnitude of hysteresis and appropriately determine the adjustment addition value during the automatic adjustment period, so that hunting due to the influence of noise included in the tension detection value Yfb can be suppressed. .
  • the control device 400 can generate a limit cycle having a constant or constant period, and can accurately calculate the value of the control gain.
  • FIG. Embodiment 5 of the control device according to the present invention will be described.
  • FIG. 10 is a configuration diagram of the control device 500 and the control target device 10 according to the fifth embodiment
  • FIG. 11 is a configuration diagram of the standard deviation estimation unit 146.
  • the standard deviation estimation unit 146 is a modification of the standard deviation estimation unit 106 of the control device 100 described above. Note that components having functions similar to those of the first embodiment are denoted by the same reference numerals as those of the first embodiment, and detailed description thereof is omitted.
  • the control device 500 includes a control calculation unit 101 that calculates an operation amount Uc, an adjustment execution command generation unit 102 that generates an adjustment execution command value SWat, a binary output unit 103 that calculates an adjustment addition value Uadd, and gain candidates And a control gain calculation unit 104 that calculates a value.
  • the control device 500 includes a control gain adjusting unit 105 that changes a gain value used in the control calculating unit 101, a standard deviation estimating unit 146 that calculates a standard deviation estimated value Sig, and a hysteresis that calculates a hysteresis width calculated value Hc.
  • a width calculation unit 107, a subtracter 108 that performs subtraction, and an adder 109 that performs addition are provided.
  • the control device 500 receives a tension command value Yr from the outside, receives a tension detection value Yfb from the control target device 10, and outputs a tension axis speed addition value Vadd to the control target device 10.
  • the standard deviation estimation unit 146 operates a high-pass filter unit 146a that calculates a low-frequency component removal signal yh, a square value calculation unit 106b that calculates an output signal y 106b by calculating a square value, and a low-pass filter. And a low-pass filter unit 106c for calculating the output signal y 106c . Further, the standard deviation estimation unit 146 includes a square root calculation unit 106d that calculates the output signal y 106d by square root calculation, a selector unit 106e that outputs the standard deviation estimated value Sig based on the adjustment execution command value SWat, A delay unit 106f that holds the standard deviation estimated value Sig.
  • the standard deviation estimator 146 calculates the standard deviation estimated value Sig based on the tension deviation value Ye during a period in which the adjustment execution command value SWat and the tension deviation value Ye are input and the adjustment execution command value SWat is off.
  • the standard deviation estimated value Sig is output.
  • the standard deviation estimation unit 146 operates so that the previous standard deviation estimated value Sig ⁇ is output in the automatic adjustment period in which the adjustment execution command value SWat is on. Therefore, the standard deviation estimated value Sig output from the standard deviation estimating unit 146 during the automatic adjustment period becomes the previous standard deviation estimated value Sig ⁇ .
  • the high-pass filter unit 146a receives the tension deviation value Ye and applies a high-pass filter to the tension detection value Ye to obtain a low-frequency component removal signal yh obtained by removing the low-frequency component of the tension deviation value Ye. Calculate and output the low frequency component removal signal yh.
  • the standard deviation estimation unit 146 estimates the standard deviation estimated value Sig that is an estimated value of the standard deviation of the low frequency component removal signal yh from which the low frequency component of the tension deviation value Ye is removed during the period when the adjustment execution command value SWat is off. Is output.
  • control device 500 sets the hysteresis width setting value Hs of the binary output unit 103 to an appropriate value, and determines the adjustment addition value Uadd based on the tension deviation value Ye and the hysteresis width setting value Hs.
  • the influence of noise included in the tension detection value Yfb is reduced.
  • the control device 500 can generate a limit cycle having a constant or constant period, and can adjust the control gain with high accuracy.
  • the control device 500 substitutes the calculated hysteresis width calculation value Hc for the hysteresis width setting value Hs.
  • the hysteresis width calculation value Hc is calculated based on the standard deviation estimated value Sig.
  • the standard deviation estimated value Sig is a good estimated value of the standard deviation of the noise signal included in the tension detection value Yfb by the standard deviation estimating unit 146 described later. That is, since the control device 500 estimates the standard deviation value of the noise signal included in the tension detection value Yfb based on the standard deviation estimated value Sig, it is possible to estimate the distribution of the noise signal generated during the automatic adjustment period.
  • the hysteresis width calculation value Hc having a magnitude that exceeds the amplitude of the noise signal can be calculated with high probability.
  • control device 500 and the control device 100 described above differ only in the configuration of the standard deviation estimation unit.
  • operations and effects when the standard deviation estimation unit 146 calculates the standard deviation estimated value Sig will be described.
  • the tension deviation value Ye input to the standard deviation estimation unit 146 is obtained by subtracting a constant tension command value Yr from the tension detection value Yfb, and only the low frequency component is different from the tension detection value Yfb. Therefore, the signal from which the low frequency component of the tension deviation value Ye is removed matches or falls within a predetermined range with the signal from which the low frequency component has been removed from the tension detection value Yfb.
  • the standard deviation estimation unit 146 is the same as the standard deviation estimation unit 106 of the control device 100 except for the high-pass filter unit 146a that calculates the low-frequency component removal signal yh, and thus has the same effect as the standard deviation estimation unit 106. can get.
  • control device 500 can set an appropriate magnitude of hysteresis and appropriately determine the adjustment addition value during the automatic adjustment period, so that hunting due to the influence of noise included in the tension detection value Yfb can be suppressed. .
  • the control device 500 can generate a limit cycle having a constant or constant period, and can accurately calculate the value of the control gain.
  • FIG. FIG. 12 is a configuration diagram of the control device 600 and the control target device 30 according to the sixth embodiment. Note that, among the components of the control device 600, components having the same functions as those of the control device 100 according to the first embodiment are denoted by the same reference numerals as those of the first embodiment, and detailed description thereof is omitted.
  • the control target device 30 includes a water tank 31 that stores the liquid 35, a temperature detector 32 that detects the temperature of the liquid 35, a heater 33 that heats the liquid 35, and a current supply device 34 that supplies current. And a liquid heating device that adjusts the temperature of the liquid 35.
  • the control target device 30 receives the current command value Ir from the control device 600 and outputs the temperature detection value Tfb.
  • the temperature detector 32 detects the temperature of the liquid 35 and outputs a temperature detection value Tfb that is a detected value.
  • the temperature detection value Tfb is a control amount, and is a value controlled so as to approach the command value as will be described later.
  • the heater 33 is supplied to the heater 33 from the current supplier 34.
  • the heater 33 generates heat based on the magnitude of the supplied current.
  • the heat generated in the heater 33 is transmitted to the liquid 35 and heats the liquid 35.
  • the current supplier 34 supplies the heater 33 with an amount of current that matches the current command value Ir based on the current command value Ir.
  • Liquid 35 is a liquid such as water, oil, or chemical liquid.
  • the temperature of the liquid 35 is increased by transferring heat from the heater 33. That is, when the value of the current command value Ir is changed, the temperature of the liquid 35 changes.
  • control target device 30 measures the temperature of the liquid 35 with the temperature detector 32 and outputs a temperature detection value Tfb. That is, the control target device 30 is configured to perform feedback control by being combined with the control device 600 that calculates the current command value Ir using the temperature detection value Tfb.
  • the control device 600 includes a control calculation unit 601 that calculates an operation amount Uc, an adjustment execution command generation unit 102 that generates an adjustment execution command value SWat that is a command value indicating whether or not adjustment can be executed, and the operation amount Uc during adjustment.
  • a binary output unit 603 that calculates an adjustment added value Uadd that is a value to be added, and a control gain calculation unit 604 that calculates a gain candidate value.
  • the control device 600 includes a control gain adjusting unit 105 that changes a gain value used in the control calculating unit 601, a standard deviation estimating unit 606 that calculates a standard deviation estimated value Sig, and a hysteresis that calculates a hysteresis width calculated value Hc.
  • a width calculation unit 107, a subtracter 608 that performs subtraction, and an adder 609 that performs addition are provided.
  • the control device 600 receives a temperature command value Tr from the outside, receives a temperature detection value Tfb from the control target device 30, and outputs a current command value Ir to the control target device 30.
  • the control calculation unit 601, the binary output unit 603, and the control gain calculation unit 604 are the control calculation unit 101, the binary output unit of the first embodiment, except that one of the input signals is the temperature deviation value Te. 103, which has the same function as the control gain calculation unit 104, a detailed description of the operation is omitted.
  • the standard deviation estimation unit 606 is the standard deviation of the first embodiment except that one of the input signals is the temperature deviation value Te and the other of the input signals is the temperature detection value Tfb. Since it has the same function as the estimation unit 106, detailed description of the operation is omitted.
  • the subtractor 608 receives the temperature command value Tr and the temperature detection value Tfb, calculates the temperature deviation value Te from the difference between the temperature command value Tr and the temperature detection value Tfb, and outputs the temperature deviation value Te.
  • the adder 609 receives the operation amount Uc and the adjustment addition value Uadd, adds the operation amount Uc and the adjustment addition value Uadd, calculates the current command value Ir, and outputs the current command value Ir.
  • the control target device 30 When the adjustment execution command SWat is off, the control target device 30 is feedback-controlled so that the temperature detection value Tfb approaches the temperature command value Tr by the operation amount Uc calculated by the control calculation unit 601.
  • the adjustment execution command SWat is on and the hysteresis width set value Hs is appropriately set, the current command value Ir and the temperature deviation value Te are limited cycle vibrations with a period that is considered to be constant or constant. appear.
  • control device 600 according to the sixth embodiment has the same configuration as that of the control device 100 according to the first embodiment, the same effect as that produced by the control device 100 can be obtained.
  • control device 600 can set an appropriate magnitude of hysteresis and appropriately determine the adjustment addition value Uadd, so that hunting due to the influence of noise included in the temperature detection value Tfb can be suppressed.
  • the control device 600 can generate a limit cycle having a constant or constant period, and can accurately calculate the value of the control gain.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)
PCT/JP2015/073055 2014-10-09 2015-08-17 制御装置及び制御方法 WO2016056305A1 (ja)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US15/508,684 US10088813B2 (en) 2014-10-09 2015-08-17 Control apparatus and control method
CN201580054433.8A CN106796416B (zh) 2014-10-09 2015-08-17 控制装置以及控制方法
JP2016552857A JP6494648B2 (ja) 2014-10-09 2015-08-17 制御装置及び制御方法
TW104132803A TWI563356B (en) 2014-10-09 2015-10-06 Control device and control method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2014208146 2014-10-09
JP2014-208146 2014-10-09

Publications (1)

Publication Number Publication Date
WO2016056305A1 true WO2016056305A1 (ja) 2016-04-14

Family

ID=55652930

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2015/073055 WO2016056305A1 (ja) 2014-10-09 2015-08-17 制御装置及び制御方法

Country Status (5)

Country Link
US (1) US10088813B2 (zh)
JP (1) JP6494648B2 (zh)
CN (1) CN106796416B (zh)
TW (1) TWI563356B (zh)
WO (1) WO2016056305A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6996624B2 (ja) * 2018-06-12 2022-01-17 東芝三菱電機産業システム株式会社 鉄鋼プラント制御装置

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS61279901A (ja) * 1985-06-05 1986-12-10 Chino Corp 調節計
JP2013235307A (ja) * 2012-05-02 2013-11-21 Mitsubishi Electric Corp 電動機制御装置

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS57199004A (en) * 1981-06-01 1982-12-06 Toshiba Corp Sample value adaptive process controller
JPH02213903A (ja) 1989-02-15 1990-08-27 Omron Tateisi Electron Co 制御装置
US5875109A (en) * 1995-05-24 1999-02-23 Johnson Service Company Adaptive flow controller for use with a flow control system
FR2765746B1 (fr) * 1997-07-01 1999-09-17 Ecole Superieure Atlantique D Procede et dispositif de commande de commutateurs pour regulation par modulation d'impulsions a frequence commandable
US6826369B1 (en) * 1999-04-23 2004-11-30 System To Asic, Inc. Intelligent sensor platform
JP2002176791A (ja) * 2000-09-26 2002-06-21 Yaskawa Electric Corp 電動機制御装置
DE10394087T5 (de) * 2003-12-25 2005-12-22 Mitsubishi Denki K.K. Motorsteuereinrichtung
JP2007137022A (ja) * 2005-11-22 2007-06-07 Fujifilm Corp 熱可塑性樹脂フィルム及びその製造方法
US8509926B2 (en) * 2005-12-05 2013-08-13 Fisher-Rosemount Systems, Inc. Self-diagnostic process control loop for a process plant
JP2007193664A (ja) 2006-01-20 2007-08-02 Nagoya Institute Of Technology リミットサイクルを用いた制御対象の動特性推定方法および制御器パラメータの設定方法
CN101229525B (zh) * 2008-02-27 2010-06-02 东南大学 雷蒙磨粉碎自动控制方法及其装置
TW200943318A (en) * 2008-03-06 2009-10-16 Nat University Corp Yokohama Nat University Control apparatus, positioning apparatus, control method, and measuring apparatus
US8508330B1 (en) * 2009-05-25 2013-08-13 Cypress Semiconductor Corporation Adaptive filter for lighting assembly control signals
US9122262B2 (en) * 2011-10-13 2015-09-01 Mitsubishi Electric Corporation Servo control device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS61279901A (ja) * 1985-06-05 1986-12-10 Chino Corp 調節計
JP2013235307A (ja) * 2012-05-02 2013-11-21 Mitsubishi Electric Corp 電動機制御装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
NOBUHIDE SUDA, PID CONTROL, 1 March 1997 (1997-03-01), pages 162 - 167 *

Also Published As

Publication number Publication date
US10088813B2 (en) 2018-10-02
JP6494648B2 (ja) 2019-04-03
CN106796416B (zh) 2019-11-05
TW201629651A (zh) 2016-08-16
US20170277143A1 (en) 2017-09-28
CN106796416A (zh) 2017-05-31
JPWO2016056305A1 (ja) 2017-04-27
TWI563356B (en) 2016-12-21

Similar Documents

Publication Publication Date Title
JP6165332B2 (ja) ロール間搬送制御装置
JP6125046B2 (ja) ロール間搬送制御装置
JP6154435B2 (ja) 制御系のオンライン自動調整状況を表示する機能を有するサーボ制御装置
JP5889497B1 (ja) ローラ間搬送制御装置
JP2016092935A (ja) ゲイン自動調整支援装置
CN109275353B (zh) 电动机控制装置
JP6494648B2 (ja) 制御装置及び制御方法
JP5614319B2 (ja) ピンチロールの速度制御システム
Wang et al. No‐Tension Sensor Closed‐Loop Control Method with Adaptive PI Parameters for Two‐Motor Winding System
Magura et al. Modeling and analysis of multi-motor drive properties in a web processing continuous line
JP2014117787A (ja) 制御装置
US20160274567A1 (en) Control device and control method
JP2010120047A (ja) 圧延機間張力制御方法及び圧延機間張力制御装置
JP2018503529A (ja) ベルト鋳造設備用のプロセス最適化
JP6256656B2 (ja) ロール間搬送制御装置およびロール間搬送制御方法
JP2008181356A (ja) 温度制御装置
JP5865138B2 (ja) 制御パラメータの決定方法及び装置
JP2013257279A (ja) 疲労試験機の振動制御装置
JP6416820B2 (ja) 制御系を自律的に安定化して自動調整を行う機能を有するサーボ制御装置
KR100584620B1 (ko) 프린터 급지 시스템의 제어기 자동 튜닝 방법 및 장치
JP2010214458A (ja) プロセスラインの張力制御方法および張力制御装置
WO2014190993A1 (en) Process control method
JP2010049307A (ja) むだ時間同定装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15849202

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2016552857

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 15508684

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15849202

Country of ref document: EP

Kind code of ref document: A1